DESCRIPTION (adapted from the Abstract): The importance of genetic variation of HIV-1 for the pathogenesis of AIDS is unclear. The Principal Investigator notes that a persistent viral infection, a high viral load, and an error-prone replication mechanism provide ideal conditions for evolutionary forces to shape the virus-host interaction during the course of infection; he speculates that this process may have importance for viral pathogenesis either through the emergence of strains capable of occupying new ecological niches, such as different host cell types, or through the generation of a population of viruses whose antigenic diversity overwhelms the capacity of the host's immune system to contain them. The Investigator and his associates have completed recently a study aimed at characterizing the in vivo evolution of plasma HIV-1 env C2V3 sequences in relation to disease progression in HIV-l infected infants. Six infants were followed with sampling at approximately six-month intervals for two years. Two progressed to AIDS within this time frame. The results from this study indicate that infants who developed AIDS had an increased rate of nucleotide substitution and showed a greater overall genetic diversity. Infants with AIDS also showed sequentially dominant lineages, suggesting a process of sequential replacement of variants, perhaps as a result of periodic escape from the immune response or due to increasing tropism for a productive host cell type. Phylogenetic analysis of deduced amino acid sequences showed a clustering of V3 loop sequences detected at later time points from the two infants with AIDS. These V3 loop sequences shared four amino acids in positions shown to be genetic determinants of T-cell line tropism. The evolutionary pattern suggests that viruses from these infants with AIDS have acquired independently, through what appears to be an example of convergent evolution, a similar and possibly more virulent phenotype. The aims of the proposed research are to test the generality of these findings. One hundred infants will be followed at three-month intervals for two years. The Investigator proposes to sample 50 sequences from reverse transcribed viral RNA at each time interval for each patient. However, by using heteroduplex analysis, he will need to sequence only a fraction of these sequences. The Investigator proposes to collect nucleotide sequence data and data for four clinical phenotypes: (1) viral load, (2) CD4+ T-cell count, (3) clinical classification, and (4) time. The data will be analyzed within a phylogenetic framework using the intraspecific cladogram estimation procedure of Templeton, which was developed for reconstruction of within-species phylogenies and allows for a nested statistical design to test hypotheses of association. The hypotheses to be tested fall into two categories: (1) genetic diversity associated with disease progression; and (2) association of genetic diversity with clinical phenotypes. In the first category the researchers will ask whether infants who rapidly progress to AIDS as compared to slow progressors have an increased rate of nucleotide substitution, a greater proportion of nonsynonymous to synonymous nucleotide substitutions, and greater HIV-l nucleotide diversity, and whether the former infants show sequentially dominant lineages and convergent or divergent motifs in the V3 loop. In the second category they will ask whether an association exists between genetic diversity and viral load, CD4+ T cell count, clinical classification or time and, if an association is found, whether the genetic avenues to the genotype related to the change in phenotype are independent or convergent.